Table 4

Estimates of the Effect of In Utero and Infant Coverage on Adult Hospitalizations, NIS 1998—2011

 All Visits Excluding Pregnancy-Related VisitsChronic Condition Related VisitsType of Chronic Condition VisitMental Health Related Visit
Diabetes/Obesity RelatedHigh Blood Pressure/Heart Related
(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)
OLSPrenatal eligibility−0.186*
(0.0939)
−0.0391
(0.104)
−0.156*
(0.0826)
−0.550**
(0.236)
−0.287
(0.244)
−0.399*
(0.237)
−0.829***
(0.269)
−0.643**
(0.274)
−0.711**
(0.292)
−0.410*
(0.248)
−0.407*
(0.239)
−0.271
(0.384)
−0.136
(0.177)
−0.108
(0.201)
0.00713
(0.121)
IVPrenatal eligibility−0.237**
(0.109)
−0.184
(0.119)
−0.260**
(0.114)
−0.766**
(0.321)
−0.560*
(0.328)
−0.555*
(0.310)
−1.034***
(0.394)
−0.958***
(0.353)
−0.840**
(0.392)
−0.323
(0.324)
−0.227
(0.338)
−0.0309
(0.464)
−0.0268
(0.190)
−0.0824
(0.231)
0.0337
(0.144)
Region × birth year fixed effects X  X  X  X  X 
State-specific birth year trends  X  X  X  X  X
Incidence (per 10,000 individuals) 374.0  25.3  15.5  9.8  77.7 
N 3,527  2,836  2,653  2,221  2,643 
  • Notes: This table displays regression results using the 1998–2011 Nationwide Inpatient Sample discharges excluding cases where the primary diagnosis is related to pregnancy or delivery. Robust standard errors clustered by state are in parentheses. Dependent variable is the log of the number of visits by category for each state–year–birth cohort. States are excluded if there are zero discharges for any state–year–birth cohort observation. All models include state by year and birth year fixed effects, and state by birth year control variables (see text). Additionally, region by birth year fixed effects and state-specific linear trends in birth year are included where indicated. First stage for IV model is reported in Online Appendix Table A.4. Significance levels:

  • * p < 0.1,

  • ** p < 0.05,

  • *** p < 0.01.